ABSTRACT Frontotemporal dementia (FTD) is a common neurodegenerative cause of an early age-of-onset dementia. Changes in personality, social, and emotional function characterize the behavioral variant of FTD (bvFTD) and since there is partial overlap with the symptoms of psychiatric illness patients often receive early diagnoses of major depressive disorder (MDD) or bipolar illness (BPAD). A delay in receiving the correct diagnosis negatively impacts the care these patients receive. 10-40% of FTD is inherited, most commonly by autosomal dominant mutations in one of three genes (MAPT, GRN, and C9orf72). By studying the earliest behavior changes in individuals who carry one of these FTD mutations we can identify features that distinguish mood disorders from FTD. Reward processing involves a determination of what an individual finds pleasant and will pursue or work for. Many of the symptoms in bvFTD reflect a shift in reward processing, including changes in motivation for food, sex, alcohol, money, and social approval. Patients with bvFTD have been shown to be particularly insensitive to things that others would find negative or aversive. The proposed research will compare reward processing in presymptomatic patients with genetic FTD and those with mood disorders. We will study 60 patients with presymptomatic genetic FTD, 60 gene negative members of the same families, 40 patients with bvFTD, 40 with MDD, 40 with BPAD, and will compare them with 60 healthy controls. The central hypothesis of this proposal is that reward processing differs in characteristic ways between bvFTD, even in its early stages, and mood disorders in a way that reflects the vulnerable anatomy of these disorders. In Aim 1 we will identify differences in reward processing abnormalities that distinguish bvFTD from mood disorders using a series of laboratory-based paradigms during which we will record patient responses and measure their autonomic nervous system reactivity to rewarding stimuli. In Aim 2 we will assess the diagnostic accuracy of the reward measures and classification approach from Aim 1 in separating early bvFTD from mood disorders. In Aim 3 we will correlate patients' reward task performance with the severity of their mood and behavioral symptoms and will identify the neuroanatomic correlates through structural and functional imaging. The results of the proposed research will improve early diagnosis of bvFTD through objective, laboratory based measures, resulting in better clinical care and facilitation of clinical trials; suggest reward-based targets for symptomatic therapies and reward-based measures of behavior to apply in FTD animal models. It will also expand the understanding of reward behaviors and their anatomic correlates in psychiatric illness, allowing for more targeted therapies.